Leon Derczynski
banner
leonderczynski.bsky.social
Leon Derczynski
@leonderczynski.bsky.social
LLM Security at NVIDIA

Prof in CS/NLP at IT University of Copenhagen

garak guy, garak.ai

"berømt skikkelse"
"like a gazelle"

Copenhagen/Seattle
😂 arXiv is cute when it pretends to have standards!
November 3, 2025 at 5:53 AM
Come to LLMSEC at ACL & hear Niloofar's keynote

"What does it mean for agentic AI to preserve privacy?" - Niloofar Mireshghallah, Meta/CMU

(Friday 1st Aug, 11.00; Austria Center Vienna Hall B)

See you there!

#acl2025 #acl2025nlp
July 28, 2025 at 3:19 PM
3² + 4² = 5² ? big if true
May 21, 2025 at 10:24 AM
Here's my "Most Inappropriate Demo" trophy at NVIDIA, 2024. For garak's "atkgen.Tox" probe, an unfettered LLM used to goad other LLMs into being toxic.
March 19, 2025 at 1:30 PM
it's a weekday where I dont have to take pacific time calls
February 17, 2025 at 5:35 PM
you know the field has changed when the foreign event you were speaking at is on the tv news on the bus home
February 13, 2025 at 9:18 AM
Will be representing NVIDIA at the EU AI Summit in Paris. I'll be talking about how we build & help others build safe, secure AI systems.

On 11.2 you can see me at:

* AI Assurance and Testing: Global Perspectives

* Building trustworthy AI: balancing innovation, responsibility, and democratization
February 8, 2025 at 11:35 AM
Should've seen it coming
February 2, 2025 at 6:41 AM
it was too difficult to not buy
January 13, 2025 at 5:36 AM
Are there people who don't make the sponge cake rice cooker recipe asap?!
January 5, 2025 at 5:07 PM
Good Christmas times, finally the elephant has come to our house!
December 28, 2024 at 12:58 PM
Sokath, his eyes uncovered
December 23, 2024 at 12:49 PM
wdym. how could anyone get that impression
December 4, 2024 at 5:44 PM
see you at #chr2024! (i swore off hashtags but i guess sometimes it's hard to stay on the wagon)
December 4, 2024 at 8:44 AM
Did not expect Russian Ryanair to get this excited about Black Friday
November 29, 2024 at 8:14 AM
i suppose the fact that my reaction to this at 7.something a.m. is one of mild panic and dejection is a clue that i should not work in product or sales
November 27, 2024 at 6:49 AM
and the tests work in practice. ANSI terminal control sequence attack success rates on <a recent model> - way into the double digits!
November 25, 2024 at 3:49 PM
it turns out LLMs can output the control codes needed for ANSI control codes to run - so the entirety of those command sets is available usable through LLM output. why are these things in the tokeniser lol
November 25, 2024 at 3:49 PM
STÖK did a cool talk (www.youtube.com/watch?v=3T2A...) showing how viewing e.g. a log entry can cause code to be run on your computer without your intervention, great for hacking. all of OSC8 + OSC52 is available without user intervention. but wait, there's more!
November 25, 2024 at 3:49 PM
just found a hilarious llm vuln. you know those ANSI escape codes for colouring your terminal? they can also move the cursor, hide text, execute commands, etc. through just viewing a piece of text
November 25, 2024 at 3:49 PM
For example
November 24, 2024 at 6:19 AM
> trying to persuade my partner to get onto bsky after they already left twitter
> they can't find a pfp they want
> "just take a selfie now"
> "in the bathroom? no. never. bathroom selfies are utter trash. the worst"
> "lol 🫠"
November 23, 2024 at 3:04 PM
Here's the French minister of AI, the convening Columbia prof, and the president of Mozilla, standing on an office table in a bar on a swaying boat
November 21, 2024 at 1:27 PM
Back in snowy Denmark. I always forget how good the showers tend to be here - in heat, control, water quantity, and bathroom layout.
November 21, 2024 at 9:55 AM